LONDON (Reuters) – Bank of England bond buying may actually increase the speed with which future interest rate rises make UK government debt unbearable and is one glimpse into why central banks need to tread lightly around post-pandemic debt piles for many years.
Britain publicly prides itself on keeping the average maturity of its national debt much longer than other major economic powers – sensibly reducing the threat of refinancing crunches, rollover risks and cutting short-term interest rate sensitivity into the bargain.
The UK Treasury’s Debt Management Office claims the average maturity of its total stock of debt was a whopping 14 years at the end of last year – well over twice the U.S. equivalent and six years longer than the next closest G7 partner France.
But after warning about the interest rate sensitivity of UK government debt for the past year, the Office for Budget Responsibility (OBR) – the government’s own budget watchdog – last month detailed just why the seemingly comfortable G7 comparison disguised a counterintuitive effect of the BoE’s bond buying.
The Bank of England’s (BoE) on-off government bond buying programmes since the financial crash 12 years ago, at least partly designed to keep long-term borrowing rates low, basically involve the purchase of gilts from banks in return for interest-bearing reserves at the central bank rather than cash per se.
And the interest rate on those bank reserves – currently a record low 0.1% – is the very ‘Bank Rate’ the BoE uses to adjust its overall monetary policy and against which so much other household and corporate borrowing rates are referenced.
The OBR’s point is that while the average maturity of outstanding government debt may indeed be 14 years, the length of net debt of the public sector as a whole – including bank reserves held at the BoE – has plummeted as those bank reserves have ballooned over the past decade or more.
By the end of next year, the OBR estimated that some 32% of gross government debt – or 875 billion pounds ($1.2 trillion) – will be held in the form of bank reserves.
The initial saving for government is obvious – replacing gilts with an average interest rate of 2.1% with bank reserves offering just 0.1% and bagging a net saving of almost 18 billion pounds for the current financial year in the process.
But the price is replacing bonds maturing in 13 years on average with overnight liabilities that would move up in lockstep with a rise in the BoE’s policy rate.
“This dramatically increases the sensitivity of debt interest spending to changes in short-term interest rates,” the OBR report warned.
With the BoE’s latest quantitative easing programme still underway, the effect is increasing.
LESS THAN A YEAR
Already the median maturity is three years lower than the headline average of 14 years and then taking account of the BoE purchases and bank reserves cuts that median to just four years. Adding another several months of BoE buying and including government liabilities overall – such as Treasury bills and savings products – then the median maturity will plummet to less than one year by March 2022.
The sharp end of the OBR’s conclusion is that if interest rates and bond yields were 100 basis points higher than assumed over the next five years, debt interest spending would be almost 21 billion pounds higher in 2025-26 – which would be 0.8% of national output and two thirds of the fiscal tightening announced by the Treasury in March.
It cushions that by saying if the reason for higher interest rates is much faster growth and higher tax receipts, then the cost will be bearable. But if rising rates are due mainly to a rising risk premium, debt sustainability could be threatened.
The counterview is whether some reform of interest on bank reserves could address the problem. Already the U.S. Federal Reserve adjusts interest on bank reserves independently of its main policy rate and the European Central Bank and Bank of Japan have some form of tiering of rates too.
William Allen at independent UK think tank the National Institute of Economic and Social Research said possible remedies being discussed by leading economists include making bank reserves non-interest bearing while changing BoE policy rate signalling to securities repurchase markets.
But this would just act as a tax on banks, costing them about 800 million pounds a year at current rates or many times more if rates rose, and potentially raise bank funding questions, hurt their stock valuations and sow financial stability issues. It may also complicate monetary policy transmission.
Instead, Allen suggested a compulsory swap of banks’ reserve balances for short and medium-dated fixed-rate gilts.
“If the proposal were adopted, the government would have more time to make fiscal adjustments in the event of a rise in interest rates,” he wrote in a policy paper published March 26.
Others stress the BoE already favours first selling its gilts back to the market as a form of tightening before any rate rise and this could neutralise the problem – even if it ends up keeping base rates lower for a very long time.
Whatever the likely fix, it shows how difficult a negative interest rate proposal was and is for the BoE, as well as how fragile pandemic-related public debt surges are when central banks are still trying to meet strict inflation targets while retaining operational independence.
More broadly it underlines the constraint on all central banks in ‘normalising’ interest rates once the pandemic is over.
(By Mike Dolan, Twitter: @reutersMikeD. Additional reporting by Andy Bruce.; Editing by Susan Fenton)